Job Description

In the early 1900's Benjamin Holt replaced the wheels on a steam tractor with wooden tracks bolted to chains in order to allow the tractor to work in more rugged conditions. Watching nearby, a bystander remarked that the machine crawled along much like a caterpillar, and thus Caterpillar's history began. In roughly 90 years, Caterpillar has transformed the world as a pioneer in the mining, construction, and energy application spaces, leading in development and innovation of disruptive technology. Caterpillar has grown into a world class manufacturer with over $53 billion in sales and revenue in 2019.

Caterpillar is the leader in industrial, autonomous earthmoving solutions and we’re looking to grow our team as we develop technology to meet our customers’ strenuous requirements. Are you tired of effortlessly solving your 17x17x17 Rubik's cube? Are Millennium Prize Problems a little too easy? Join us.  Our newly established Cat Robotics team is ushering automation & autonomy technology into the rugged production environment found on mining and construction jobsites. You will be joining an elite team with a startup feel, while maintaining access to a vast network of experienced professionals.

As an Autonomy Perception Engineer, you’ll develop algorithms that enable autonomous systems to understand the world and navigate safely. That includes detecting, tracking, and predicting the motion of pedestrians and vehicles, as well as characterizing obstacles and terrain. We’re looking for a primarily hands-on engineer, who’s capable of taking real-world problems, turning them into well-defined projects, surveying and selecting the right approach, developing quick prototypes, bringing them to production, and mentoring jr. engineers.

Job Duties

Be a part of the Perception team on our autonomous vehicle platform. 

Research, prototype, and bring to production perception algorithms that enable autonomous and semi-autonomous systems to understand the world around them and navigate safely

Design and implement computer vision systems on an autonomous vehicle platform

Work with data from various perception sensors including Lidar, Camera, and Radar

Work with simulation and AV system testing teams to guide simulation sensors and perception evaluation in support of the perception stack

Optimize on-board perception code

Design metrics and architect software to gauge performance against requirements

Mentor and grow more junior members while interacting with senior technical leaders

Basic Qualifications

Bachelors or Master’s degree in Robotics, Computer Science, or Engineering

5+ years relevant work experience

Industry experience implementing perception algorithms for detection, tracking, segmentation, pose estimation. 

Experience with a variety of sensors including LIDAR, stereo/mono cameras, radar, and IMUs

Experience with real-time sensor fusion (e.g. IMU, lidar, camera, odometry, radar)

Industry experience building optimized perception pipelines using ROS, C/C++, OpenCV and/or CUDA

Excellent C++ coding, strong engineering practices, debugging/profiling skills

Strong foundation in mathematics and fundamentals (3D geometry, linear algebra)

Ability to convert research papers into production implementation.

Top Candidates will also have

Knowledge of robotics and frameworks such as ROS

Experience with SLAM, filtering, and state estimation techniques

Experience with machine learning and classification, exposure to deep learning frameworks

Experience working with GPUs, particularly on embedded hardware

Publications in top-tier computer vision/robotics conference (CVPR, ICCV, ECCV, ICRA, IROS)

Additional Information

The location for this position is San Francisco, CA

Domestic Relocation assistance is available for this position

This position may require 10% travel

Visa sponsorship is available for this position